Skip to content
Makes the ads

AI video18 min read

How Long Should Video Ads Be? Lengths by Platform, Goal and Product (2026)

How Long Should Video Ads Be

How long should video ads be? For most social placements, start between 6 and 30 seconds: 6 seconds for reminders, 15 seconds for a single clear offer, and 30 seconds when the product needs a demo. Go to 60 seconds or longer only when the viewer chose to keep watching, as on YouTube skippable ads or retargeting. There is no single magic number, because the right length depends on three things: the placement, the campaign objective and how much your product needs explaining.

This guide gives you the official limits for Meta, TikTok, YouTube, LinkedIn, Snapchat and connected TV, the starting ranges we use on live campaigns, a plan to cut one shoot into four lengths, and a simple way to test length so your own data picks the winner. We also cover something most guides skip: Arabic and English versions of the same ad rarely fit the same number of seconds.

In this guide

The Short Answer

If you need one rule, use this: make the ad as short as the message allows, then let the placement set the ceiling. A video ad is long enough when it has a hook, one clear idea, proof and a call to action. Anything past that is cost without benefit.

Here is the quick version we hand to clients before a brief:

  • Reminder or brand recall: Starting length: 6 seconds; Why: One image, one line, the logo
  • Single offer or launch: Starting length: 15 seconds; Why: Hook, product, offer, call to action
  • Product demo or comparison: Starting length: 30 seconds; Why: Room to show the product working
  • Considered purchase or B2B: Starting length: 60 seconds or more; Why: Only where viewers can skip or have chosen to watch

Treat these as starting points, not laws. A 45-second founder story can beat a 15-second cut if the first 3 seconds earn attention. A 6-second bumper can beat both for recall. The rest of this guide explains how to decide, and how to prove it.

Platform Limits vs Recommendations vs What Wins

Most advice on ad length mixes up three different numbers. Separating them saves a lot of confused briefs.

1. The hard limit. This is the longest video the platform will accept. Limits are often huge. Meta's Facebook Feed accepts video from 1 second to 241 minutes, according to Meta's video ad specs. Nobody should run a four-hour ad, so the hard limit rarely matters.

2. The playback rule. This is what the placement does to your video, and it matters far more than the limit. Instagram Stories plays ads under 16 seconds in full but may split longer ones into separate cards. Facebook in-stream shows only the first 15 seconds of a longer ad. LinkedIn loops videos under 30 seconds until they reach 30 seconds of playback. YouTube lets viewers skip after 5 seconds. These rules decide how much of your ad people actually see.

3. The length that wins. This is the length that hits your goal at the best cost, and only testing tells you. Platform recommendations are a sensible place to start, but they average across millions of advertisers who sell different things to different people.

When you brief a video, write down all three for each placement. It stops the classic mistake of producing one 45-second edit and discovering that half the placements cut it or skip it.

Video Ad Length by Platform

Below are the limits from each platform's own help pages, checked in October 2026, plus the starting range we use. Specs change often, so confirm them in your ads manager before you export.

Meta: Feed, Stories and Reels

Meta's limits are generous. Facebook Feed takes 1 second to 241 minutes. Instagram Reels ads take 0 seconds to 15 minutes. The general technical limit for Stories is 1 second to 60 minutes.

The playback rule is what shapes our edits. Instagram Stories plays video ads under 16 seconds for their full length. At 16 seconds or longer, the ad may be split into up to three cards, and some objectives show a "Keep Watching" option. So for Stories we cut to 15 seconds or less, which keeps the story in one piece.

Our starting ranges:

  • Feed: 15-30 seconds. People scroll fast, but feed viewers will stay for a strong demo.
  • Stories: 6-15 seconds, built vertically for 9:16.
  • Reels: 15-30 seconds, edited to feel like native Reels content rather than a TV spot.

For safe zones, aspect ratios and how to plan one shoot for every Meta placement, see our Meta ads creative size guide.

Facebook In-Stream

In-stream ads run inside videos people chose to watch, so the viewer is waiting for their content to return. Meta lists desktop in-stream at 5 to 15 seconds and mobile at 5 seconds to 10 minutes. Ads of 15 seconds or less play in full and may be skippable after the first 5 seconds. Longer ads show only their first 15 seconds, with the rest available through a card below the video.

In practice, that makes 15 seconds the real ceiling. Our starting range is 5-15 seconds, with the brand and the offer inside the first 5.

TikTok

TikTok's specs page says non-Spark in-feed ads can run up to 10 minutes, and Spark Ads (which boost an organic post) have no length restriction. The useful number is in TikTok's own creative guidance, which recommends 21-34 seconds for in-feed ads while noting that shorter videos generally perform better.

We start TikTok tests at 15-34 seconds. The first 2 seconds carry the hook, and the edit should look like a TikTok, not a commercial that wandered onto the app. If your ads stall before you can even read retention data, our tour of TikTok Ads Manager settings covers the delivery and bidding checks to run before you blame the edit length.

YouTube: Bumper, Non-Skippable, Skippable and Shorts

YouTube is the platform where length rules differ most by format. Google's YouTube video ad formats page sets them out:

  • Bumper ads: 6 seconds, non-skippable. Built for reach and recall.
  • Non-skippable in-stream: Google now lists 15-60 seconds depending on the campaign subtype. Most buys still use 15 seconds.
  • Skippable in-stream: viewers can skip after 5 seconds. There is no max length, but Google recommends under 3 minutes.
  • In-feed: no max length, because the viewer clicks a thumbnail to watch.
  • Shorts: Google recommends under 60 seconds, and viewers can swipe away at once.

The skippable format changes the length question. You pay attention with the first 5 seconds, and everyone who stays past the skip button chose to be there. That is why a 60-second or longer skippable ad can work for a considered product: the viewers who stay are the ones you want. Our starting range for skippable is 15-30 seconds for direct response and 30-60 seconds for products that need a demo.

For how to build each of these formats so the first 5 seconds earn the view, see our format-by-format YouTube creative guide.

LinkedIn

LinkedIn accepts standard video ads from 3 seconds to 30 minutes, and in-stream video ads (in gradual testing) up to 90 seconds. The LinkedIn video ad specs recommend 15-30 seconds so the video qualifies for every placement across the LinkedIn feed and Audience Network. Videos under 30 seconds loop until they reach 30 seconds of playback.

B2B buyers will watch longer when the content is useful, but the feed is still a feed. We start at 15-30 seconds for awareness and test 60 seconds only for thought-leadership clips or demos to warm audiences. Our sibling post on LinkedIn video ads best practices goes deeper on hooks and targeting for B2B.

Snapchat

Snapchat single video ads can run up to 3 minutes, and they stay skippable the whole time. Snapchat's own guidance has long pushed advertisers toward very short creative, because the viewer is tapping through friends' stories. Check the current spec in Snap's ads manager before export, since it changes.

Our starting range is 5-10 seconds, vertical, sound on, with the product in the first frame. Gulf audiences use Snapchat heavily, so for UAE and Saudi campaigns we often build this cut first rather than last.

Connected TV

Connected TV (smart TVs and streaming boxes) is the one place where traditional TV slot lengths still rule. Spots are usually sold as 15 or 30 seconds, and the viewer sees the whole ad on a big screen, usually with sound. YouTube non-skippable ads also run on TV screens.

On CTV, length is less about attention and more about pacing. A 30-second spot gives you room for a story, but remember there is no click. Put a clear brand cue in the final 5 seconds and a short web address or search prompt viewers can remember.

How Objective and Product Change the Answer

Placement sets the ceiling. Your objective and your product decide where under that ceiling you land.

Awareness and Reach

For awareness, you are buying memory. Short wins here because you want the brand seen and remembered by as many people as possible. We use 6-second and 15-second cuts, with the brand shown early, and run them at high frequency. A 6-second bumper with one striking visual will often do more for recall than a 30-second story most people never finish.

Consideration and Traffic

Consideration is where 15-30 seconds earns its keep. The viewer is curious but not convinced. You need a hook, a demo and a reason to believe. If you sell something new to the category, lean toward 30 seconds. Our guide to short-form video ads breaks down how to structure that middle section so it does not sag.

Conversion and Retargeting

People who visited your site, watched your earlier ads or added to cart already know you. For them, a longer ad can answer the objections that stopped them: delivery times, sizing, warranty, how it works. We test 30-60 seconds in retargeting, but also very short 6-15 second "reminder plus offer" cuts. Both can win, depending on why people dropped off.

Product Complexity

The more your product needs explaining, the longer the ad, up to a point. A perfume sells on mood and identity, so 15 seconds of texture, light and a name can be plenty; our perfume and oud video ads work leans on short, sensory cuts for that reason. A commercial van, an SUV or software needs proof: size, features, a walkthrough. For a car dealer client selling commercial vans and SUVs, we made filmed and AI video ads for each model, each focused on one vehicle so the length could go to the details buyers ask about. That campaign brought in more than a hundred leads in the first two weeks, with about one in four reaching the qualified stage.

If you are planning a demo-heavy ad, our sibling post on product demo video examples shows how strong demos pace the "show it working" moment.

One Shoot, Four Lengths: The Cut-Down Plan

The biggest length mistake is producing one edit. Platforms, objectives and audiences need different lengths, so plan them before the shoot, not after. One shoot, four lengths:

  1. Write the 60-second version first. It holds the full story: hook, problem, demo, proof, offer, call to action. Every shorter cut is a subset of it.
  2. Mark the 30-second version. Keep the hook, one demo moment, one proof point and the offer. Drop secondary features and extra testimonials.
  3. Mark the 15-second version. Hook, product in use, offer, call to action. One idea only.
  4. Mark the 6-second version. One striking shot, the product name and the logo. No argument, just memory.
  5. Shoot for all four at once. Capture extra close-ups, a clean product hero shot and vertical framing so each cut has the footage it needs.

Two production notes make this work. First, frame for vertical and horizontal on the day, or shoot with enough room to crop both, because the 15-second cut will mostly run in 9:16 and the CTV cut in 16:9. Second, record lines as standalone sentences. A line that depends on the sentence before it cannot survive a cut.

This is also how we keep cost down for clients without cutting quality: one production day becomes a full set of placement-ready lengths. If you want a team to plan that from the brief, our video production services cover strategy, filming, AI production and editing.

What Each Length Must Keep

Every version, even the 6-second one, needs three things: a reason to stop scrolling in the first seconds, the brand clearly shown, and one takeaway. If a cut loses any of those, it is not a shorter ad, it is a broken one. When you review cut-downs, mute them and watch them once at full speed. If you cannot tell what is being sold, recut.

Arabic and English Versions Run at Different Speeds

Here is a gap almost every length guide misses. If you advertise in the UAE, Saudi Arabia or the wider Gulf, you are often running Arabic and English versions of the same ad, and they do not fit the same seconds.

Arabic and English scripts differ in length and rhythm. A line that takes 4 seconds in English can take more or less time in Arabic, depending on dialect, phrasing and the voice. Arabic captions also read right to left and need their own space in the frame. If you translate an English 15-second script line by line and force it into 15 seconds, the voice-over ends up rushed or the captions fall off before people can read them.

What we do instead:

  • Write the Arabic script first when Arabic is the main audience, then adapt to English. Our Arabic video production is built on this: scripts, voice and captions in Arabic first, never dubbed.
  • Time each language separately. The Arabic 15-second cut may need fewer words, not faster speech.
  • Keep the visual timeline shared where you can, so both versions use the same shots and only the voice and text change.
  • Test each language as its own ad. Retention curves often differ, and so does the winning length.

For auto brands in the region, this is common: a dealer may run an Arabic cut to local buyers and an English cut to expat buyers. Our car dealer video ads are planned bilingual from the brief for that reason.

How to Test Video Ad Length

Opinions about length are cheap. A clean test is cheap too, if you run it properly. Here is the method we use.

  1. Pick one placement and one objective. Testing a 15-second TikTok cut against a 30-second YouTube ad tells you nothing about length.
  2. Use the same creative at different lengths. The 15-second and 30-second versions must share the same hook and offer. If you change the hook, you are testing hooks, not length.
  3. Run them side by side in the same ad set or a proper A/B test, with the same audience and budget, so delivery is not skewed.
  4. Set a decision metric before launch. Pick cost per result for conversion campaigns, cost per ThruPlay or cost per completed view for awareness, and cost per landing page view for traffic.
  5. Wait for enough data. Give each version time to exit learning and gather a meaningful number of results before you call it. Early winners often reverse.
  6. Read the retention curve, not just the result. If the 30-second version wins on results but loses most people at second 8, recut its middle and test again.
  7. Keep the winner, then test the next variable. Length is one lever. Hook, offer and format still move results more often than seconds do.

A small ad-spend example: if each version gets the same daily budget for a week, you can compare cost per result directly. Keep the budget identical and the audience identical, or the test is noise.

When a winning length starts to slip after a few weeks, it may not be the length at all. Our explainer on what creative fatigue is shows how to tell tired creative from a wrong length.

Retention Metrics That Tell You the Right Length

Every major ads manager reports how long people watch. These are the numbers we read to decide length.

Hook rate (3-second views divided by impressions). This shows if the opening stops the scroll. A low hook rate is an opening problem, not a length problem. Fix it before you shorten anything.

Hold rate (ThruPlays or 15-second views divided by 3-second views). This shows if the middle keeps people. If the hook is strong but hold is weak, the ad is probably too long for what it says. Cut the slow middle.

Video quartiles (25%, 50%, 75% and 100% watched). These show where people leave. A sharp drop at 25% points to a weak transition after the hook. A steady fade is normal. A cliff right before the offer means the offer arrives too late, so move it earlier.

ThruPlay and completed views. On Meta, a ThruPlay counts when the video plays to completion, or for at least 15 seconds if it is longer. Compare cost per ThruPlay across lengths to see which delivers more full views for the money.

View rate on YouTube. For skippable ads, view rate shows how many people stayed past the skip button. If view rate is fine but conversions are weak, the ad may be holding attention without moving people to act.

Click-through and conversion rate. Retention is a means to an end. A shorter cut with weaker completion can still win if it drives more clicks.

Read these together. One metric alone can mislead you into cutting a good ad short, or keeping a long one alive because the completion rate looks fine.

Common Video Ad Length Mistakes

We see the same mistakes again and again when we audit ad accounts.

  • One edit for every placement. A 45-second feed ad cut off at 15 seconds in in-stream and split into cards in Stories.
  • Burying the brand. If the logo or product only appears in the last 5 seconds, most viewers never see it.
  • Trimming instead of rewriting. Cutting a 30-second ad to 15 seconds by chopping the end removes the offer and the call to action.
  • Slow intros. Logos, fades and scene-setting in the first 3 seconds cost the most valuable time in the ad.
  • Translating, not adapting. Arabic and English versions forced into the same timeline.
  • Calling a length test too early. A two-day result is a guess, not a finding.
  • Ignoring sound-off viewing. Without captions, a 30-second talking ad is 30 seconds of silent faces.

A practical example from our side: a Dubai restaurant and catering client wanted event bookings, not likes, and their best ads had started to lose effectiveness. Instead of making new long ads, we relaunched their four best videos with a form that asked for event type, number of guests and date. The result was more than a hundred event enquiries in under two weeks of the relaunch. The lesson for length: before you produce more seconds, check whether the ads you have just need a sharper ending and a better next step.

If you want a broader quality check before you worry about seconds, our 10-point video ad scorecard covers hook, proof, offer and call to action.

Frequently Asked Questions

What is the best length for a video ad?

For most social placements, the best starting length is 15 seconds: long enough for a hook, the product, an offer and a call to action. Use 6 seconds for reminders and recall, 30 seconds for demos, and 60 seconds or more only where viewers can skip or have chosen to watch, such as YouTube skippable ads or retargeting. Then test, because the winning length depends on your product and audience.

How long should a TikTok ad be?

TikTok's creative guidance recommends 21-34 seconds for in-feed ads, while noting that shorter videos generally perform better. Non-Spark in-feed ads can run up to 10 minutes, and Spark Ads have no length restriction. We start tests at 15-34 seconds, put the hook in the first 2 seconds, and edit so the ad feels native to the app rather than like a TV spot.

How long should a YouTube ad be?

It depends on the format. Bumper ads are 6 seconds and cannot be skipped. Non-skippable in-stream ads are usually 15 seconds, with some campaign subtypes allowing up to 60. Skippable ads have no max length, though Google recommends under 3 minutes, and viewers can skip after 5 seconds. Shorts ads should stay under 60 seconds. Put your brand and key message in the first 5 seconds.

Are shorter video ads always better?

No. Shorter ads usually get higher completion rates, but completion is not the goal on its own. A 30-second or 60-second ad can win when the product needs a demo, when the audience is already warm, or when the format lets viewers choose to keep watching. The right length is the shortest one that still delivers the hook, the proof and the call to action.

Should Arabic and English versions of an ad be the same length?

Not always. Arabic and English scripts differ in length and rhythm, and Arabic captions need their own reading time and space. Forcing a translated script into the same seconds usually makes the voice rushed or the captions unreadable. Write and time each language separately, keep the visuals shared where possible, and test each version as its own ad, because the winning length can differ by language.

How do I know if my video ad is too long?

Check the retention curve in your ads manager. If the hook rate is healthy but the hold rate is weak, or the quartiles show a sharp drop well before the offer, the ad is probably too long for what it says. Cut the slow middle, move the offer earlier, and test the shorter version against the original with the same hook, audience and budget.

Get Every Length From One Brief

The answer to "how long should video ads be" is not one number. It is a set of lengths, planned from one brief and one shoot, matched to each placement and tested against real results. Start with 6, 15 and 30 seconds for social, add 60 seconds where viewers choose to watch, and let hook rate, hold rate and cost per result tell you which one to scale.

If you want a team to plan the cut-downs, write bilingual Arabic and English scripts, and deliver every length ready for Meta, TikTok, YouTube, LinkedIn and Snapchat, book a strategy call with XMA and get a quote for your next campaign. We are based in Dubai and work with brands across the Gulf and globally.

Want this for your brand?

Tell us what you sell. We'll tell you what we'd make first.